Ventura Labs

In this episode of the Ventura Labs Podcast, we’re joined by Micaela Bazo and Pedro Penna to dive deep into Subnet 68 and its groundbreaking drug discovery capabilities. Micaela and Pedro discuss how Subnet 68 harnesses billions of molecules in chemical universes and the innovative methods miners use to search, predict affinities, and prioritize synthesizable compounds for targets like behavioral drugs from platforms powered by Bittensor's decentralized network.They explore the powerful incentive functions of Subnet 68, where miners are rewarded only for efficient, actionable shortlisting that advances novel drug candidates, and how this ensures the accuracy and translatability of predictions across preclinical stages.We also dive into the exciting partnerships Subnet 68 has formed with other entities, including collaborations with contract research organizations (CROs) and joint ventures like DiaGen AI and Contango Labs. These partnerships leverage Subnet 68's unique ability to provide both AI-generated molecule libraries and wet-lab validation for applications ranging from longevity therapies to mental health interventions.The conversation touches on the development of an open drug screening platform, the future potential of multimodal translational models, and the balance between geographical arbitrage and regulatory efficiency in biotech.Micaela and Pedro also share their thoughts on how Subnet 68 is creating value for AI training and fine-tuning, using its massive computational swarm to fuel Metanova's development. We also discuss the challenges and opportunities in data privacy, monetization, and creating miner-driven datasets that participants can share and monetize.

In this episode of the Ventura Labs Podcast, we’re joined by Will Squires and Chris Zacharia to dive deep into Subnet 13 and its groundbreaking data scraping capabilities. Will and Chris discuss how Subnet 13 has reached a staggering 55 billion rows of data and the innovative methods miners use to scrape and validate social media content from platforms like X (formerly Twitter), Reddit, and YouTube. They explore the powerful incentive functions of Subnet 13, where miners are rewarded only for real, unaltered data, and how this ensures the accuracy and reliability of the data.

We also dive into the exciting partnerships Subnet 13 has formed with other subnets, including collaborations with 57 Gaia and Score Subnet 44. These partnerships leverage Subnet 13's unique ability to provide both historical and real-time data for applications ranging from weather forecasting to sports performance analysis. The conversation touches on the development of a data marketplace, the future potential of multimodal data, and the balance between scale and speed in data processing.

Will and Chris also share their thoughts on how Subnet 13 is creating value for AI training and fine-tuning, using its massive dataset to fuel IOTA’s development. We also discuss the challenges and opportunities in data privacy, monetization, and creating user-driven data sets that miners can share and monetize.

In this episode of the Ventura Labs Podcast, we sit down with James Woodman, COO of Targon, and Joshua Brown, the CTO of Manifold Labs, to explore the cutting-edge world of decentralized compute and Bittensor subnets. They break down the inner workings of Targon’s architecture, including how validators, miner proxies, and CVMs interact within the system. We also dive into the mechanics of Targon’s auction system, emission pools, and the role of collateral in improving compute quality for enterprise clients.

James and Josh discuss the vision behind Targon, focusing on creating transparent, scalable compute markets and the long-term impact on the AI industry. With insights into the future of decentralized infrastructure, market liquidity, and fine-tuning services, this episode provides a comprehensive overview of how Targon is shaping the future of decentralized computing.

We sit down with Siam Kidd, founder of DSV—the first hedge fund dedicated to Bittensor—to explore why decentralized AI is poised to outpace today’s centralized models.

In this episode you’ll learn:

• Why targeting untapped “virgin landscapes” uncovers the highest-growth opportunities
• How subnets operate as standalone AI ventures under the Bittensor umbrella
• The island-infrastructure analogy for building out energy, data, and compute layers
• Siam’s multi-step framework for rigorously evaluating subnet potential
• The role of emissions patterns, tokenomics, and holder concentration in investment decisions
• How the 128-subnet (dTAO) cap consolidates ecosystem energy and boosts performance
• Why moonshot projects like Templar and IOTA will deliver 90 % of Bittensor’s long-term value

Key takeaways:

• Decentralized AI is the future frontier beyond “just crypto.”
• First movers build the roads; later entrants reap the greatest rewards.
• A strong evaluation process is crucial for identifying high-value subnets.
• Infrastructure—from energy generation to model deployment—underpins every layer of value.

Sound bites:

• “Bittensor isn’t really crypto — it’s fundamentally about scalable, decentralized AI.”
• “Virgin landscapes are where money flows.”
• “With a 128-subnet cap, we ensure more ‘cylinders are firing’ across the network.”

Stay tuned for Part 2

In this episode, Ventura sits down with Wouter Haringhuizen, the weather savant behind Subnet 18 on Bittensor. They dive into how Subnet 18 became a world-first, decentralized weather-forecasting network that delivers 30% more accurate temperature predictions than traditional models.

You’ll learn:

• How miners are incentivized with dTAO-backed rewards tied to real API revenue
• The origin story: from mining crew to B2B weather API startup
• Building strong miner relationships and collaborating with Bitmind
• Why focusing on core forecasts (temp, rain, wind) is the fastest path to market
• Real-world use cases in energy grids, hedge funds, and beyond
• The role of the Alpha token, revenue-sharing, and long-term utility
• Plans for expanding into carbon credits, specialized regional models, and more
• Practical advice for anyone launching a new subnet or decentralized service

Whether you’re interested in DeSci, Web3 APIs, or the future of climate tech, this episode offers a front-row seat to building a high-performance, permissionless forecasting platform.

In this episode, Ventura sits down with Steffen Cruz, CTO & Co-Founder of Macrocosmos, and Felix Quinque, AI Lead, to pull back the curtain on building a world-first, fully decentralized pipeline-parallel training architecture atop Bittensor. They trace Subnet 9’s evolution from an isolated Kaggle-style model to an IOTA-powered, layer-group system that slashes entry costs and accelerates LLM training.

You’ll learn how their custom orchestrator schedules work across miners, why they spot-check with “shadow audits” and are building CLASP for lightweight validation, how 128× activation compression and butterfly all-reduce keep data flowing, and what it takes to pull off a zero-to-one breakthrough in AI. Along the way, they share lessons learned, practical advice for aspiring subnet teams, and their vision for a future where anyone with a laptop can help train billion-parameter models.

In this episode, Ventura sits down with Thomas, Taoshi’s Director of R&D, and Jordan Bonilla, Lead Data Engineer, to pull back the curtain on building a real-world trading platform atop Bittensor. They trace their journeys from early contributors on price-prediction subnets to architects of the Proprietary Trading Network (PTN), explaining how—and why—they pivoted to simulated trading and iteratively hardened PTN’s incentive mechanism against exploits.

You’ll learn how they designed and implemented Glitch Financial—a non-custodial, enterprise-grade SaaS that pipes top traders’ signals directly into users’ brokerage accounts—complete with collateral smart contracts, a 90% initial token burn, and the upcoming Active Validator Program. Along the way they reflect on handling community FUD, collaborating with ecosystem partners like Timeless, and the power of showing up in person at conferences.

• Why they joined Taoshi and what keeps them motivated
• The technical and emotional challenges of launching and scaling a subnet
• How Glitch Financial bridges open-source blockchain signals with traditional finance
• Practical advice for aspiring subnet developers on Bittensor

Tune in for a deep dive into decentralized trading, tokenomics, and the people driving innovation on the frontier of AI and crypto.

Meet Gareth, a 20-year video-industry veteran who co-founded Nigeria’s first VOD platform and now leads AI-driven upscaling and compression on the Bittensor network. In this in-depth conversation, Gareth recounts his journey from tape-to-cloud operations to building Subnet 85, and explains exactly how decentralized AI is redefining video.

• Two decades of operations in news, animation, music and sports post-production
• Lessons learned launching Iroko TV in a mobile-first, infrastructure-challenged market
• The lunch conversation that sparked Subnet 85 and rapid prototype evolution
• How AI upscaling predicts and fills pixel gaps to turn 2K archives into 4K quality
• Why AI-based compression can cut storage and bandwidth by over 50 % while preserving picture clarity
• Strategies for protecting privacy, including off-chain processing and on-chain encryption
• Addressing crypto skepticism through transparency and industry partnerships
• Building robust incentives: combining VMAF and PI-APP to guard against gaming
• The power of 256 dedicated ML engineers working on a shared subnet
• Revenue model: enterprise B2B deals with a light consumer demo to showcase the tech
• A vision for decentralized streaming backends on edge networks
• The subnet owner’s dual role in product design, market strategy, and reinvestment
• Advice for new subnet founders: plan end-to-end, nail your messaging, and prove clear value

If you’re exploring video technology, AI subnets, or crypto-driven media innovation, you’ll come away with practical insights you can apply today.

Shakeel, the dynamic subnet owner of Ridges AI, is a trailblazer in AI-powered software engineering, steering a bold new era in decentralized technology. Through Ridges AI, he’s forging a robust platform that empowers task-specialized AI agents, reigniting his passion for innovation and community-driven development in the AI landscape.

In this captivating interview, we delve into Shakeel’s meteoric rise from Agents How to Ridges AI, uncovering the strategies behind their rapid success. He shares hard-won lessons on building a resilient codebase, mastering incentive mechanisms for miners and validators, and navigating the shift to organic queries. Shakeel opens up about the commercialization of the agent marketplace, the delicate balance of leveraging private repositories, and the competitive edge of speed in saturated markets. From his personal evolution in software engineering to the transformative potential of Bittensor’s incentive-driven ecosystem, Shakeel’s raw, practical, and inspiring insights reveal how focus, community trust, and relentless execution can redefine the future of AI in software development.

Sam, the visionary subnet owner of Templar (Subnet 3) on Bittensor, is a blockchain veteran with nearly a decade of experience, now leading the charge in decentralized AI training. Through Templar, he’s building a permissionless, antifragile platform that champions freedom and sovereignty, rekindling his passion for technology’s transformative potential after years of disillusionment in the blockchain space.

In this electrifying interview, we explore Sam’s journey from a disenfranchised blockchain developer to a pivotal figure in Bittensor’s decentralized AI revolution, spotlighting the creation of Templar and its viral SoundCloud moment. He shares raw insights on Templar’s mission to master decentralized pre-training, the gritty realities of incentive mechanisms, and the critical role of miners in co-creating value. Sam also dives into the technical challenges of scaling to massive models, the importance of open-source hardware, and why Google—not OpenAI—is the true Goliath to challenge. From his unfiltered take on tokenized model ownership to his vision for Templar as an enduring ideology, Sam reveals how grit, community, and mission-driven focus can unlock Bittensor’s potential to reshape AI and empower humanity.